Hormones appear to play an important protective role in colorectal cancer (CRC) in women. The use of exogenous estrogens has been consistently associated with a roughly 40% decrease in risk of CRC, and higher levels of circulating estrogens, such as those that occur in obese women, also appear to reduce risk. It is not clear, however, whether estrogens affect risk of all types of CRC equally, or whether their effects are associated with distinct molecular characteristics and outcomes of colorectal cancers. For example, emerging evidence suggests that estrogens may be related to the presence of epigenetic changes, particularly CpG island hypermethylation, and that these epigenetic changes may affect survival after a diagnosis of CRC. We propose to extend our recently completed case-control study of postmenopausal hormones and CRC risk (CA 76366) to examine the relationship between estrogen use, the presence of specific molecular tumor characteristics, and CRC outcomes (i.e., new primary CRC and death). We will specifically evaluate: 1) the relationship of pre-diagnosis exposures, such as postmenopausal hormone (PMH) use, with CRC outcomes;2) the relationship of PMH use with the CpG island methylator phenotype (CIMP) status and the hMLH1 promoter methylation status of the colorectal tumors;and 3) the association between molecular features of the original tumor (i.e., CIMP status, hMLH1 promoter methylation, hMLH1 expression) and CRC outcomes. To accomplish these specific aims, we will analyze tumor tissue and questionnaire data from an estimated 900 of the 1,018 women aged 50-74 years with invasive CRC who were identified from the Puget Sound SEER registry and participated in our prior study. We will use follow-up information from the SEER registry to assess the occurrence of new primary CRC and vital status over an average of five years following diagnosis, and methylation will be assessed using MethyLight. This research will provide practical information for women with CRC and their physicians about the possible prognostic importance of PMH use, and it may also lead to a better understanding of the mechanisms through which hormones affect risk of CRC and specific molecular subtypes of CRC in women.